Previous studies suggested that one of the hydrolysis products of indomethacin, either 4-chlorobenzoic acid or 5-methoxy-2 methyl-3-indole acetic acid, can inhibit platlet aggregation in vivo. If correct, this hypothesis explains the apparent action of indomethacin dissolved at high pH where hydrolysis rapidly occurs. Moreover, if the indole is the active products, the hypothesis in addition suggests a possible role for the indole following the use of indomethacin dissolved at lower pH, since the indole is also an important natural metabolic od indomethacin in man. We induced platelet aggregation in cerebral and mesentric microvessels and inhibited this aggregation with the indole (25 mg/kg i.p. one hour before test). The 4-chlorobenzoic acid was inactive. The indole had a modest but statistically significant inhibitory effect in vitro, when added to platelet rich plasma stimulated to aggregate by arachidonic acid (0.5 mM). Thus the action of indomethacin dissolved at high pH is explained, and a pharmacologic action of a metabolite of indomethacin becomes a possibility.